# mcue > Pick up where the work actually left off. mcue saves what happened, what you decided, and what comes next as an explicit work record, so a fresh AI chat can continue from there without rebuilding the context. Use it in the browser with mcue Hub (invite-only alpha), where connected AI apps read and save the record, or locally with the Apache-2.0 CLI, which works offline without an account. Git records your source history; mcue preserves the state needed to continue the work. Four commands do the real work: `mcue init` registers a project, `mcue closeout` records a session, `mcue resume` returns a bounded context packet small enough to paste into an agent conversation, and `mcue index` shows the whole portfolio at a glance. State is Markdown with YAML front matter under `$MCUE_HOME` (default `~/.mcue/`). Agents can read and write the same state over MCP — locally via `mcue serve --mcp`, or through the hosted endpoint at https://api.mcue.dev/mcp. Paste https://mcue.dev/agents-snippet.md into a repository's AGENTS.md to teach an agent the access rules. ## Getting started - [Overview](https://mcue.dev/docs): What mcue does, the two ways in, and how these docs are arranged. - [Getting started](https://mcue.dev/docs/getting-started): Save and resume your work, use plans and the Hub, and follow examples for writing, research and software. - [What mcue can do](https://mcue.dev/docs/features): The complete feature map: local operating state, agents, Hub, BYO-S3 sync, and every CLI namespace. - [Install](https://mcue.dev/docs/install): Install the CLI with your alpha download key, and confirm it with mcue doctor. - [Quickstart](https://mcue.dev/docs/quickstart): Register a project, close out a session, and resume with bounded context. ## Guides - [Close out and resume](https://mcue.dev/docs/guides/closeout-and-resume): The two commands the whole product is built around, and how to write a closeout worth resuming from. - [A week with mcue](https://mcue.dev/docs/guides/a-week-with-mcue): One project, three closeouts, one interruption, and the resume that pays for all of it — every command and its real output. - [Give agents your state](https://mcue.dev/docs/guides/agents-and-mcp): Run the local MCP server and connect a client so agents read the same state you do. - [Publish to Hub](https://mcue.dev/docs/guides/publish-to-hub): Log in, publish a project, and connect the hosted MCP endpoint. - [Read continuity graphs](https://mcue.dev/docs/guides/continuity-graphs): Follow projects into their lanes, then read the cadence and coverage of accepted closeouts. - [Work offline and sync](https://mcue.dev/docs/guides/offline-and-sync): What happens when a machine is offline, and how pending work reconciles on reconnect. - [Recover a broken session](https://mcue.dev/docs/guides/recovery): Interrupted closeouts, degraded checkpoints, and retroactive capture. ## Concepts - [Operating state](https://mcue.dev/docs/concepts/operating-state): The boundary between what mcue owns and what Git owns. - [Why not a notes file?](https://mcue.dev/docs/concepts/why-not-a-notes-file): The obvious objection, answered without hedging — including the cases where a plain notes file really is the right answer. - [Projects, lanes, and plans](https://mcue.dev/docs/concepts/projects-lanes-plans): How work is partitioned, and why a lane is not a branch. - [Authority and conflicts](https://mcue.dev/docs/concepts/authority-and-conflicts): Sequences, pending operations, divergence, and why there is no last-write-wins. ## Reference - [CLI commands](https://mcue.dev/docs/reference/cli): Every command, grouped by namespace. - [Files on disk](https://mcue.dev/docs/reference/files): What lives under $MCUE_HOME and the shape of each file. - [MCP tools](https://mcue.dev/docs/reference/mcp): The tool surface exposed to agents, locally and through Hub. ## Optional - [Full documentation as one file](https://mcue.dev/llms-full.txt): every page above, concatenated. - [Agent access rules](https://mcue.dev/agents-snippet.md): the snippet to paste into AGENTS.md or CLAUDE.md.